Vehicle Trajectory Plausibility Check Using Swarm Road Boundary Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems face challenges in determining the plausibility of driving trajectories when lane recognition is not possible due to lack of recognizable lanes, and there is a need to utilize external swarm data for improved functionality.
Innovation Solution
A method that utilizes swarm data to check the plausibility of driving trajectories by comparing them to sensed roadway boundaries, employing plausibility conditions based on distance and angle thresholds, and considering the credibility of the swarm data in relation to sensor data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If driver assistance systems rely on camera data for lane recognition, then lane detection can be achieved when lanes are clearly visible, but the system becomes unavailable when lanes are not recognizable
Solution Approach 1:
The patent introduces swarm data from other vehicles as an intermediary information source. When the host vehicle's camera cannot recognize lanes, the system uses lane marking data collected and shared by surrounding vehicles to determine the driving trajectory, thereby maintaining system availability under varying road conditions
Solution Approach 2:
The system is designed to accept multiple data sources for trajectory determination: primary reliance on own vehicle's camera data when available, and fallback to swarm data from other vehicles when camera recognition fails. This multi-functional approach ensures the driver assistance system can operate across diverse road conditions and visibility scenarios
2Reliability
If swarm data from multiple vehicles is utilized to determine driving trajectory, then the driver assistance system can function when lanes are not recognizable, but the credibility and accuracy of the trajectory data becomes uncertain
Solution Approach 1:
The system performs a plausibility check by comparing the driving trajectory derived from swarm data against actual sensor data from the host vehicle's sensors. This feedback mechanism verifies whether the swarm-based trajectory aligns with real-world observations, ensuring accuracy and credibility before accepting the trajectory for navigation
Solution Approach 2:
The patent combines swarm data from multiple vehicles with real-time sensor data from the host vehicle. By merging these data sources and performing plausibility verification, the system leverages the collective information while maintaining accuracy through local validation, thus resolving the credibility concern
3Measurement precision
If plausibility checking is performed by comparing driving trajectory to sensed roadway boundary, then the accuracy of trajectory verification is improved, but the computational complexity and time required for verification increases
Solution Approach 1:
The plausibility checking process extracts and compares only the essential geometric features: the driving trajectory from swarm data and the roadway boundary from sensor data. By focusing on these key elements and their spatial relationship rather than processing all raw sensor data, the system achieves accurate verification while managing computational complexity
Data Source
AI summary
The disclosure relates to a method for checking plausibility of at least one portion a driving trajectory for a vehicle. First, the portion of the driving trajectory is provided by way of a storage unit external to the vehicle based on swarm data. For a roadway of the vehicle, at least one segment of a roadway boundary is sensed by way of a sensing unit of the vehicle during the operation of the vehicle. The at least one portion the driving trajectory is compared with the at least one segment of the roadway boundary based on a first plausibility condition. The at least one first marking portion is compared with the at least one segment of the roadway boundary based on a second plausibility condition. The plausibility of the at least one portion is checked in accordance with the comparisons.